Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

PSYC30013 - Theme 4: Psychological Assessment > W12 - Validity > Flashcards

W12 - Validity Flashcards

1
Q

What is the formal definition of validity

A

The degree to which evidence and theory suport the interpretation of test scores entailed by proposed use of tests

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is validity concerned and not concerned with?

A
  • Concerns interpretation and use of tests
    • Must be grounded in relevant evidence and theory
  • Not merely conerned witth test items
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Is validity continous/binary

A

Continous (Like reliability)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What does validity tend to require

A

Tend to require reference to external criterion

  • Not a must
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the classic tripartite distinction

A
  1. Criterion Validity
  2. Content Validity
  3. Construct Validity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is criterion validity. What are the types.

A

The extent it can predict scores on relevant criterion variables

  • Concurrent validity
    • Test scores evaluated against criterion measured at same time
  • Predictive validity
    • Test scores evaluated against criterion measured at later time
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the typical criterion validity in practice

A

Validity Coefficient

  • Pearson correlation (typically) between test scores and scores on criterion measure
    • Attenuated due to measurement error
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Why bother with criterion validity if we already have a criterion?

A
  • Gold standard test may be more expensive/difficult to apply
  • Some criterion test are post-hoc by nature
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is content validity

A

The extent its content reflects the full domain of the construct it is suppoedly assessing

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is one way of examining content validity?

A
  • Ask experts whether individual items are “essential”, “useful but not essential”, or “not necessary” to job performance
  • Calculate the Content Validity Ratio (CVR)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is face validity. And what is it related to. Is it important

A

Face Validity

  • Face validity is the extent it seems (to test-takers) to have content validity
  • Related to content validity
  • Practical Benefit - Public Relations
    • Lack of face validity = Lack of confidence in effectivness of test = Decrease in motivation for test-taker
    • This will skew the results even though a test that lacks face validity may be relevant and useful!
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is construct validity. What are the types.

A

The extent it reflects the construct it is meant to be reflecting (i.e. appropriateness of inference)

  • Convergent Validity
    • Test scores correlated with tests of identical or related construct
  • Discriminant Validity
    • Test scores uncorrelated with tests of unrelated (which should not be theoretically correlated) construct
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is Sensitivity. What is the formula (Think through diagram)

A
  • Se
    • Test’s ability to correctly detect positive cases
      • True Positive / True Positive + False Negatives
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is Specificity. What is the formula (Think through diagram)

A
  • Sp
    • Test’s ability to correctly detect negative cases
      • True Negative / True Negative + False Positive
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is Positive Predictive Power. What is the formula (Think through diagram)

A
  • PPP
    • Probability a positive test result indicate a positive case
      • True Positive / True Positive + False Positive
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is Negative Predictive Power. What is the formula (Think through diagram)

A
  • NPV
    • Probability a negative test result indicate a negative case
      • True Negative / True Negative + False Negative
17
Q

What is prevalance. And how does it related to PPP and NPVs

A
  • Prevalance
    • Probability a random case is criterion positive
      • (True positive + False negative) / All
  • Intepretations of PPV and NPV can only be applied if base rates is applied to the population
18
Q

What is a common issue in interpreting PPP and NPVs

A

In clinical settings, the prevalance of diagnosis will be higher than aforementioned population

  • This implies that with a test with less than perfect sensitivity/specifcity, we should be wary of ruling in or out a particular condition on the basis of test results
19
Q

What is a practical understanding of PPP and NPV

A
  • PPP
    • How worried you should be if your test results come positive
  • NPV
    • How happy you should be if your test results come negative
  • Note: Only appropriate if base rates are applied to me
20
Q

Spin and Snout

A
  • Spin
    • Specific test
    • postive rules in disease
  • Snout
    • SeNsitive test
    • negative rules OUT disease
21
Q

What are 5 types of evidence proposed to be relevant to construct validity

A

CAR-IT

  • Consequences of Use
    • Should not have unfair consequences
  • Associations with other variables
    • Match between actual variables correlated with and what it should correlate with (theoratically)
  • Response Processes
    • Match between psychological processses and what the item is asking
  • Internal Structure
    • If test is one thing, items should be correlated well
  • Test Content
    • Content validity
22
Q

Estimating validity: What are nomonlogical networks

A
  • Nomological networks
    • General philosohpical approach to construct validity

System of laws which constitute a theory and which relates

  • Observable properties to each other
  • Theoretical constructs to observerable
  • Theoretical constructs to each other
23
Q

What does nomological networks help to establish

A

Multitrait-Multimethod Martix

24
Q

What does multitrait-multimethod matrices do?

A

Evaluate construct validity

  • By evaluating trait and method variance, they aim to establish convergent and discriminant validity

“Establish support claims for construct validity.”

25
Q

What sources of variance does multitrait-multimethod matrices use. What is the problem?

A

Trait Variance

  • Measures tend to share variance if they’re based on same (or similar) traits

Method Variance

  • Measures tend to share variance if they’re based on the same data source

Correlations between measures could be explained by trait/method variances, or both.

26
Q

Multitrait-multimethod matrices: What are the 4 types of corelations and expected correlations

A
  • Heterotrait-Hetereomethod Correlations
    • Weakest Correlations
  • Heterotrait-Monomethod Correlations
    • Moderate Correlations?
  • Monotrait-Heteromethod Correlations
    • Moderate Corelations?
  • Monotrait-Monomethod Correlations
    • Strongest Correlations
27
Q

Multitrait-Multimethod Matrix: How do we examine it?

A

Monotrait-monomethod

  • Reliability
  • Diagonal Line
  • Highest correlation

Monotrait-Hetereomethod

  • Converging Validity
  • Second Diagonal Line
  • Moderate Correlation

Hetereotrait-Monomethod

  • Triangle with unbroken lines
  • Moderate Correlation
  • Maybe discriminant validity

Hetereotrait-Hetereomethod

  • Triangles with broken lines
  • Weakest Correlation
  • Maybe discriminant validity

PSYC30013 - Theme 4: Psychological Assessment (2 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions